Rapid communications Most but not all laboratories can detect the recently emerged Neisseria gonorrhoeae porA mutants – results from the QCMD 2013 N . gonorrhoeae external quality assessment programme

D Luijt (d.lujit@infectielab.nl)1, C Di Lorenzo2, A M van Loon2, M Unemo3 1. Certe, Department of Medical Microbiology, Groningen, the Netherlands 2. Quality Control for Molecular Diagnostics, Glasgow, Scotland, United Kingdom 3. WHO Collaborating Centre for Gonorrhoea and other STIs, Swedish Reference Laboratory for Pathogenic Neisseria, Department of Laboratory Medicine, Microbiology, Örebro University Hospital, Örebro, Sweden

We describe the results of the Quality Control for Molecular Diagnostics 2013 Neisseria gonorrhoeae external quality assessment programme that included an N. gonorrhoeae strain harbouring an N. meningitidis porA gene which causes false-negative results in molecular diagnostic assays targeting the gonococcal porA pseudogene.Enhanced awareness of the international transmission of such gonococcal strains is needed to avoid false-negative results in both inhouse and commercial molecular diagnostic assays used in laboratories worldwide, but particularly in Europe.
In recent years, false-negative test results using PCRs targeting the Neisseria gonorrhoeae porA pseudogene have been reported from Australia, Scotland, Sweden and England [1][2][3][4].Further investigations revealed that the gonococcal strains were not clonal, but all had replaced large segments or their entire N. gonorrhoeae porA pseudogene with an N. meningitidis porA gene.This report describes the results of the Quality Control for Molecular Diagnostics (QCMD) 2013 N. gonorrhoeae External Quality Assessment (EQA) programme.It included an N. gonorrhoeae strain containing an N. meningitidis porA gene which gives rise to falsenegative results in molecular diagnostic assays targeting the gonococcal porA pseudogene.QCMD (www.qcmd.org) is an independent international organisation which provides a wide range of molecular EQA services in the field of infectious diseases to over 2,000 participants in over 100 countries.

Quality assessment
The N. gonorrhoeae porA mutant strain included in the QCMD 2013 N. gonorrhoeae proficiency testing programme (NgDNA13) was isolated in 2011 in Sweden [3].The strain was cultured on gonococcal agar media and diluted in an N. gonorrhoeae-negative urine sample to a concentration of 1.0×10 4 copies/mL.Following lyophilisation, the sample was distributed on 17 June 2013 to participating laboratories as part of the QCMD NgDNA13 proficiency testing panel, along with instructions on how the samples were to be processed.Laboratories tested the panel samples using their routine molecular diagnostic method for the detection of N. gonorrhoeae.Test results, together with details of the assays used, were returned to QCMD via a dedicated online data collection system.The deadline for submitting results was 19 July 2013.

Results
The NgDNA13 panel was sent to 286 laboratories in 35 countries, 23 of which are located in the World Health Organization (WHO) European Region.A total of 304 datasets with unambiguous test results (i.e.positive or negative) for the N. gonorrhoeae porA mutant were returned (Table 1).Some laboratories used several molecular diagnostic methods and delivered more than one dataset.Participants were also requested to specify the target gene of their assay, and 281 datasets (92%) contained this information (Table 2).
In total, 27 datasets (9%) reported the N. gonorrhoeae porA mutant sample as negative, and there was a highly significant association between reporting use of the porA pseudogene as target gene and reporting negative results in the N. gonorrhoeae molecular diagnostics (chi-square test, p<0.001).In total, 29 datasets reported the porA pseudogene as their sole assay target.Of these 29 datasets, 18 reported a negative result and 11 reported a positive result.The additional nine datasets that could not detect the N. gonorrhoeae porA mutant reported as target the pivNG gene (n=2), porA and 16S rRNA gene (n=2), porA and opa genes (n=1), opa genes (n=1), 16S rRNA gene (n=1), cryptic plasmid and amino acetyltransferase gene (n=1), or did not report the gene (n=1) (Table 2).
Results in 268 datasets (88%) were from a commercially available molecular technology.In 36 datasets (12%), an in-house PCR assay had been used.Of the most frequently used commercially available N. gonorrhoeae molecular assays, only one manufacturer (Seegene) was reported to use the N. gonorrhoeae porA pseudogene as the only target gene.Fourteen datasets (5%) were generated by Seegene assays (Table 2) used in six different countries (data not shown).In addition, 14 of the 36 in-house molecular assays also used the porA pseudogene as the sole target gene.

Discussion
Gonorrhoea remains a major public health problem globally [5], and N. gonorrhoeae has developed resistance to all antimicrobials used for treatment of gonorrhoea, which is of grave concern worldwide [6].In settings with sufficient resources, molecular diagnostic methods have to a large extent replaced conventional culture diagnosis.However, a considerable number of molecular diagnostic assays for N. gonorrhoeae (both commercial and in-house) have shown cross-reactivity with other Neisseria spp.[7][8][9].This suboptimal specificity has led in Europe and Australia to the recommendation that positive tests should be confirmed with another molecular detection assay targeting a different gene sequence [10,11].The gonococcal porA pseudogene is often used as a target sequence in in-house PCRs and in some new commercial assays for confirmatory testing, and in some settings also for primary diagnostic examination.
In the present quality assessment for the detection of N. gonorrhoeae, the opa genes were the most commonly used individual targets (25%), followed by the DR-9 repeat sequence gene (19%), 16S rRNA (13%) and the pivNG gene (11%).Forty-six datasets (15%) were generated by a method targeting the N. gonorrhoeae porA pseudogene, either alone (n=29) or in combination with additional genes (n=17).Surprisingly, 11 laboratories using assays reported to target the N. gonorrhoeae porA pseudogene alone (Seegene assays and some inhouse real-time PCRs), were able to detect the N. gonorrhoeae porA mutant sample.These results require further investigation, and it is likely that additional molecular tests and/or targets were used but were not reported.For example, although the porA pseudogene was reported as the sole target by all participants, the Seegene AnyplexII STI-7 Detection and Seegene Seeplex STI Master ACE Detection assays are dual-target assays (porA pseudogene and one additional gene) according to the manufacturer.
Most molecular diagnostic assays targeted other genetic sequences than the N. gonorrhoeae porA pseudogene, and accordingly detected the porA mutant sample correctly.However, an enhanced awareness of the potential emergence of gonococcal mutants resulting in false-negative results in the molecular diagnostic tests is essential because of the ongoing evolution  and genetic diversity of N. gonorrhoeae and other bacteria.For example, the new variant of Chlamydia trachomatis reported from Sweden had a 377 bp deletion in the cryptic plasmid, which contained the target sequences for the C. trachomatis molecular assays from two main manufacturers and resulted in many thousands of false-negative tests in Sweden [12][13][14].
With the replacement of conventional culture methods, laboratories need to be aware of the risk of emergence of these mutant strains that cannot be detected by molecular assays.
In conclusion, enhanced awareness of the international transmission of N. gonorrhoeae porA mutant strains is needed to avoid false-negative results in several molecular diagnostic assays, both in-house and commercial.The opportunities to use combinations of different diagnostic methods such as several molecular methods, molecular methods and culture, and multitarget methods, remain exceedingly valuable.

Table 1
Participating laboratories in the QCMD 2013 Neisseria gonorrhoeae external quality assessment programme, 2013 (n=286) a Participants may submit more than one dataset, e.g. when they have several N. gonorrhoeae assays available.

Table 2
Results for the Neisseria gonorrhoeae porA mutant sample NgDNA quality assessment programme, 2013 (n=304) a As reported by 2013 external quality assessment participants